In order to evaluate the clinical efficacy of the fully coated HA implants, the power and the significance levels were set to 80% and 5%, respectively.

The null hypothesis was defined as having equivalent mean values of alveolar bone resorption between the fully HA-coated implant (μ₁) and the partially HA-coated implant (μ₂) groups. The alternative hypothesis was defined as having different mean values, as indicated in the following equations.

Assuming that the intergroup difference (δ) would be >0.5 mm and the variance of bone resorption (σ²) would be (0.8 mm)², for a statistical significance level of 0.05, Z_α would equal 1.96, and for a statistical power of 80%, Z_β would equal 0.84, σ² would equal 0.8, and δ would equal 0.5. The number of subjects necessary for the study was calculated as follows:

Since approximately 2 implants were placed in each subject, approximately 25 patients were needed in each group, for a total of 50 subjects.

Assuming that the dropout rate would be approximately 20%, the necessary number of subjects was N×(1/0.8)=50×1.25=approximately 64. Thus, each group was assigned 32 subjects.

A total of 75 participants (48 males and 27 females, mean age of 58.7±10.3 years) who received care at the Department of Oral Maxillofacial Surgery of the Seoul National University Bundang Hospital, Korea were screened for a randomized controlled clinical trial. The full exclusion criteria are presented in Table 1, and a total of 11 subjects were excluded for the following reasons: 9 exhibited no evidence of moderate periodontitis, 1 did not submit a consent form, and 1 did not demonstrate sufficient residual alveolar bone. Thus, 64 of the 75 recruits remained for randomization. All remaining patients had chronic periodontitis with moderate crestal bone loss around the edentulous premolar or molar regions. The severity of periodontitis was characterized on the basis of the amount of clinical attachment loss and probing depth (PD) according to the Centers for Disease Control-American Association of Periodontology definition, and only patients with moderate periodontitis were included in the study (Table 2) [13].

After randomizing 64 subjects, 11 more patients who withdrew consent (n=9) or were lost to contact (n=2) were excluded from the study. Thus, 53 patients remained in the study, and each of them was treated with one or more of the 2 types of HA-coated implants (Osstem Implant, Seoul, Korea). A total of 71 implants were placed. In group I, 36 partially HA-coated TS III implants were placed, while in group II, 35 fully HA-coated TS III implants were placed. Subsequently, 9 patients dropped out of the study due to a primary implant stability quotient (ISQ) value of <50 (n=6), sinus perforation (n=1), poor bone quality (n=1), or nonspecific pain (n=1). Ultimately, 21 patients in group I (30 implants) and 23 patients in group II (32 implants) were examined and analyzed (Figure 1A and B).

Depending on the available width of the crestal bone ridge, implants with diameters of 4.5 mm and 5.0 mm were chosen for placement. The implants were 8.5 mm and 10.0 mm in length. During the implant placement surgery, a sinus lift through the crestal approach and minor guided bone regeneration (GBR; with the use of a resorbable collagen membrane) around the implant were performed when the surgeon decided that they were necessary. The crest modules of the implants were placed 0.5 mm below the alveolar ridge, using either a 1-stage or a 2-stage technique. If more than 2 implants were placed at once, diagnostic cast models were used to pre-fabricate the surgical stent. All implants were placed with torque between 30 and 40 N∙cm. The implant-placement protocols and drilling sequence followed descriptions provided in the manufacturer’s surgical manual.

Panoramic and periapical radiographs were taken postoperatively. The sutures were removed 10–14 days after surgery. General medications, such as broad-spectrum antibiotics, nonsteroidal anti-inflammatory drugs, anti-ulcer agents, and 100 mL of 0.1% chlorhexidine via mouth gargling were prescribed for 5 days. If the implant was placed using a submerged technique, a second operation was performed and a healing abutment connection was placed 3 weeks after the first operation in the mandible and 8 weeks after the first operation in the maxilla. In both the submerged and nonsubmerged techniques, the measurement of the secondary ISQ, impression taking, and bite registration were performed 4–5 weeks after implant placement in the mandible and 10 weeks after implant placement in the maxilla. The final prostheses were typically delivered and loaded 3 months after placement in the maxilla and 6 weeks after placement in the mandible. All prosthodontic procedures were carried out by a highly experienced prosthodontist.

In order to evaluate the stability and clinical outcomes of the implant, primary and secondary ISQ values were measured before prosthesis loading using a resonance frequency analyzer (Osstel Mentor, Ostell, Göteberg, Sweden). Primary stability was measured after an implant was placed at its final depth, and secondary stability was measured immediately before impression taking for prosthesis fabrication. For precision, the mean ISQ value was calculated using values measured at the buccal, lingual, mesial, and distal aspects around a given fixture.

Peri-implant parameters, including the modified plaque index (mPI), modified sulcus bleeding index (mSBI), and probing depth (PD), were assessed at 4 aspects around the implant 12 months postoperatively [1415]. Peri-implantitis was diagnosed when the peri-implant condition satisfied all three of the following criteria: mucositis with bleeding or suppuration on probing, a PD >5 mm, and progressive crestal bone loss of >5 mm.

Panoramic and standardized periapical radiographs were taken preoperatively and postoperatively to examine the implant sites and their proximal edentulous areas. For both linear and dimensional accuracy of the intraoral radiographs, the paralleling technique using a customized plastic film holding device was used. All radiographs were obtained by 2 experienced radiological technologists who were able to provide good reproducibility.

The baseline intraoral radiographs were taken immediately after surgery, and compared with radiographs taken 12 months after the prosthesis loading in order to observe marginal bone loss differences. Radiographic examinations were inspected independently by a single clinical evaluator to identify any progressive bone loss (Figure 2A and B).

Distances between the implant shoulder and the first visible bone-implant contact (mm) were measured using PACS version 3.0.9.1 (INFINITT, Seoul, Korea). The clinician made 2 marks designating where the crestal bone intersected the implant body as shown in the software visualization. The mesial and distal bone loss of the implants were measured to calculate the mean marginal bone loss. Changes in the crestal bone height of each implant were calculated based on the differences between the initial and final measurements made from standardized periapical radiographs. The magnification rate was taken into consideration to compensate for the proportional differences between the real implant length and the length indicated on the radiographs. Each gap between the threads of the implant fixtures was 1 mm, and this known implant dimension was used as a dimensional reference in evaluating the alveolar bone loss observed in radiographs. Measurements were rounded to the nearest 0.01 mm.

The success criteria for the osseointegrated dental implants were as follows: (1) no persistent or irreversible subjective pain or complaints, (2) no recurring peri-implantitis infection or suppuration, (3) no perceptible mobility, and (4) no continuous radiolucency around the implant-bone junction [1516]_. An implant was also considered a failure if more than 1 mm of bone resorption was found 1 year after the loading (or 0.2 mm/year afterward). If the implants were not physically removed, however, failed implants were included in the survival rate, but not in the success rate [17].

In order to measure overall satisfaction levels with the dental implant therapy, all patients filled out questionnaires after the final implant prosthodontic loadings. Satisfaction levels were rated on an 11-point scale, where 10 was the most positive response (completely satisfied) and 0 was the least positive response (completely dissatisfied). The mean overall satisfaction levels were used to evaluate patients’ overall satisfactions with both types of implants.

The Wilcoxon signed-rank test and Mann-Whitney test in SPSS version 19.0 (SPSS Inc., Chicago, IL, USA) were used to evaluate differences between and within the 2 groups. P values <0.05 were considered to indicate statistical significance.